Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
Coal & It's Utilization Short 1986
COAL AND ITS UTILIZATION SHORT COURSE October .27, 1986 School of Mineral Engineering University of Alaska - Fairbanks CHAPTER 6 CONVENTIONAL AND ADVANCED CLEAN COAL USE TECHNOLOGIES INTRODUCTION The combustion of coal to generate steam, heat, and electricity is presently a subject of great interest to industries which have previously used oi] and gas. A common interest among current and prospective users, whether generated by corporate policy, public demand or government law, is the availability of cost effective coal combustion systems which are environmentally sound and which are capable of meeting the energy requirements of an industrial or utility plant. In most industrial plants, the cost of energy is a substantial factor in the cost of the finished product. Also, in most energy producing systems, the fuel cost is the largest single cost component in the life cycle of the total system. Therefore, it is important to select combustion systems that offer the greatest choice and fuel flexibility over the life of the operating system. In this chapter # brief overview is presented on conventional and advanced methods of burning coal. The general technical characteristics, design parameters and relative economic advantages are highlighted for each-combustion system. 1 1 This chapter is largely based on material from papers by Ronald C. Lutwen and R.A. Malone presented at the First Annual Fluidized Bed Conference in December, 1985 sponsored by thé Council of Industrial Boiler Owners. 115 DIRECT COAL COMBUSTION SYSTEMS The coal combustion systems covered in this chapter are spreader stokers, pulverized coal boilers (PC), and fluidized bed boilers (FBC). Each system’s operating parameters are generally described. These descriptions are followed by a discussion on the comparative differences among the systems. See schematics of these systems at end of this chatper. Spreader Stoker The spreader stoker is the industry’s standard coal combustion system. It has been used extensively in industrial applications burning a wide range of coals and waste fuels. Spreader stokers are designed to permit continuous fuel feed, stable fuel burning, reasonable response to load variations, and easy removal of the residue ash from the storage pit. Fuel is evenly distributed over the width of the travelling grate at a volume to match the boiler heat load. The grate travels from the back of the furnace to the front keeping the fuel and ash bed at a depth of two to four inches. Combustion is completed at the front and the ash is dumped into the ash pit. Some designs of spreader stoker units provide for a recirculation of part of the ash to reduce the amount of unburned coal, Combustion in spreader stokers is a combination of fine particles burning in suspension and larger particles burning in the bed. Generally, high pressure overfire air jets above the bed generate turbulence to assure complete combustion of the fine particles. Depending upon the air emission requirements, the combustion gases are scrubbed of particulate matter. 116 Pulverized Coal Combustor A pulverized coal combustor (PC) burns finely powdered coal in suspension in a water tube boiler. The coal is ground to 70 percent minus 200 mesh and blown into the boiler with excess air volume ranging from 20 to 30 percent. Hot primary air is used to blow the pulverized coal into the boiler. Secondary air for complete combustion is supplied by forced draft fans. As the hot combustion gasses flow upward they are cooled by the furnace heat transfer systems. Before the combustion gases can be discharged into the atmosphere, they must go through various air quality control equipment dependent upon emission standards. Atmospheric Fluidized Bed A fluidized bed combustor (FBC) is composed of coal and a bed material, either sand or a sulfur sorbent such as limestone, contained in a vessel. The bed is "fluidized" by air flowing from beneath the vessel with sufficient velocity to suspend the bed and result in turbulent mixing. This turbulence causes high heat transfer between the bed and combustion gases with the boiler’s heat transfer systems, and encourages capture of the sulfur dioxide by the sorbent. Before discharge of the combustion gases into the atmosphere, the entrained particulate matter must be removed. There are two types of atmospheric fluidized bed combustors, the bubbling bed (BBC) and the circulating bed (CBC). In the bubbling bed the fluidizing air velocities suspend the bed, but do not transport the bed out of the combustion chamber. In circulating beds the air velocity is high enough to expel the bed from the combustion chamber. The entrained solids are captured 117 in cyclone separators and returned to the combustor. COMPARATIVE ADVANTAGES OF DIRECT COAL-FIRING SYSTEMS In this section the principal characteristics and comparative differences among the three coal combustor systems are discussed. Capacity Ranges Stoker systems are supplied primarily to industrial markets in size ranges from 10,000 to 400,000 pounds of steam per hour. Pulverized units have been designed with steaming capacities as low as 60,000 pounds per hour for small industrial plants and up to 10 million pounds per hour for electric utility plants. Fluid bed units range from 10,000 to 650,000 pounds of steam per hour for industrial systems, and up to 925,000 pounds of steam per hour for utility systems. Tennessee Valley Authority (TVA) is building a 160 megawatt unit with a maximum continous boiler steaming rate over 1.1 million pounds per hour. The choice of a coal combustion system is an economic decision and should only be made after a thorough and detailed study of load requirements, available coal supplies, and environmental restraints. For smaller installations, particularly those needing only low pressure saturated steam, consideration should be given to purchase of standardized, packaged designs that are shipped assembled. Care must be taken that the system, will be able to use the type of coal that is planned. Larger installations are generally custom designed, and therefore offer more opportunities for optimization both for load requirements and fuel choice than do packaged units. The choice 118 of system should weigh heavily the longer term economics and not be driven by the initial capital investment. Coal Sizing and Preparation Coal size requirements differ widely for the three combustion systems. The greater degree of coal treatment required, the higher the cost for the coal feed. Justification for a washed coal (to lower both sulfur and ash content) is both an economic and environmentally driven decision. Of the three combustion systems, the pulverized unit requires the least coal sizing, since the crusher at the boiler prepares the final product going into the furnace. Generally, only a top size is specified to permit uniform feeding to the crusher. The circulating fluid bed requires the next level of coal sizing; usually a product sized to one-half to three-quarters inches is satisfactory. The bubbling fluid bed requires a more narrow size range, generally with a topsize of. three-eights inches. Coal dryers are typically required for bubbling bed units. Stoker units require the most exact size specifications with limits on both the top and fine sizes. Stoker coal is usually double screened and washed. Some users of stokers buy run-of- mine coal and do their own crushing and sizing. Coal Characteristics All three combustion systems are reasonably tolerant to a wide range in the quality of the coal feed. However, there are 119 some special considerations relevant to each of the three systems. Pulverized coal systems are sensitive to several coal characteristics, such as hardness, slagging, fouling, and moisture. For proper and efficient operations, PC units must be designed for the specific coal to be used. Fluid bed systems will burn almost every type of coal, including low rank coals of poor quality. Although there are few limitations, the main consideration is that a reduction in the Btu content of the coal correspondingly affects the heat output. Stoker systems will burn a wide range of coals, from lignite to bituminous, without regard to the caking and fouling characteristics of the coals. Slagging may be a problem if the unit is operated above the fusion temperature of the coal. Boiler Efficiency Boiler efficiency is a function of three main factors: unburned coal, excess air, and exit flue gas temperature. The net effect of these three factors gives PC units the highest potential efficiency rating, with FBC and stoker units typically five percentage points less efficient. From practical experience, the actual day to day operating efficiencies are usually less than the stated manufacturer’s rating. As shown in Table 6.1 the typical values for excess air are lower for PC and FBC units than stoker and that excess air varies from full load to minimum load. The unburned coal for PC units is lower than that for the other two systems. However, the level of unburned combustibles can be varied by increasing the amount of ash reinjection. Flue gas temperatures can be similar for all 120: types of units since this depends on the heat transfer equipment used. Table 6.1 COMPARATIVE EFFICIENCY VALUES * (in percent) STOKER Pc Excess Air 100% Rating 30 20 50% Rating 50 35 Minimum Load 60 40 Unburned Combustible Range 2-7 -5 -. Typical 4 5 Unit Efficiency Range 80 - 85 86 - Typical Value 84 87 Expected Oper. 82 86 88 BFB CFB 20 20 40 40 45 ; 45 2 - 10 2-4 4-8 3 - 4 80 - 83 82 - 85 82 84 80 84 * Data from Lutwen, see reference. 121 Turndown The turndown ratio is an important measure of the ability of a combustion system to follow changes in load demand. The typical load response for the three basic combustion systems is shown in Table 6.2. TABLE 6.2 TURNDOWN RATIO AND LOAD RESPONSE* STOKER Pc BFB CFB Typical Turndown 321 6:1 3:1 3:1 Load Response % 10-15 20-25 10 . 10 Time Unit minutes minutes hour minutes * Data from Lutwen, see reference. Load response for stoker systems is done by varying the amount of fuel and air. The limitations are usually set by the ability to keep a stable, non-smoking fuel bed which is partly a function of the type of fuel used. Load response is accomplished in PC units by varying the feed rate of both fuel and air. The rate of response to changing load demand is limited by the ability to maintain a stable flame and the minimum velocity of air to keep the coal in suspension in the feed pipes. There are four principal means to follow load demand with fluidized bed combustors: velocity of air, recycle rate, bed 122 mass, and bed slumping. By reducing the velocity of air, heat tubes in the bed are uncovered and there is a corresponding reduction in heat output. Increasing the ash recycle lowers the bed temperature and correspondingly the heat output. Raising or lowering the bed mass will change the heat output, but it requires considerable time and care to insure that the temperature is maintained within acceptable limits for sulfur capture. Slumping involves the removal of a complete bed from service. Although slumping gives a rapid response to load changes, it takes substantial time to bring the bed back into service. Substantial progress has been made in improving the load response of fuidized bed combustors from the early models installed by various industries. Some manufacturers offer units with a guaranteed turndown rate. Control Systems Control systems for stokers are much simpler than those required for either the pulverized or fluid bed units. However, stokers do need constant operator attention to ensure efficient operation and prevent malfunction. Complex controls are available for stokers which improve efficiencies, and these should be considered in any new installation. Pulverized coal units use automatic burner control systems in addition to combustion analyzers to maintain a stable flame. These controls are usually designed for both automatic start-up and shut-down. 123 Fluid bed combustors require sophisticated controls to monitor fuel feed, bed temperature, and other operating parameters to avoid operational problems and to control load response. Auxiliary Systems Power requirements vary for the three systems. Stokers have the lowest demand. Pulverized units require more auxiliary power, and fluidized bed even more because of the large air volumes needed to suspend the bed and circulate the combustion gasses. Auxiliary systems and equipment associated with the three combustion type units vary, and further depend upon the size of the unit. The major common items are condenser, circulating water system, cooling tower, feedwater heaters, boiler feed pumps, and feedwater treatment. Auxiliary systems specific to the type of combustion unit are building size, ash handling, coal handling, fans, instruments and controls, electrical equipment, piping, and wiring. For these items overall costs are lowest for stokers, with PC units having higher costs, and FBC the highest. Air Emission Control The three primary air pollutants from the combustion of coal are particulates, sulfur oxides, and nitrogen oxides. The basic control methods for each of these pollutants are addressed in this section. Particulate control is standard and proven technology whether using electrostatic precipatators or baghouses. The 124 choice is not only a function of installed capital and maintenance costs, but flexibility to accommodate a range of coals. Reducing sulfur oxide emissions from stoker and pulverized combustor systems requires the use of flue gas scrubbers. There are two basic scrubber options, wet systems requiring sludge disposal, and spray dryer systems which produce an easily disposed of harmless dry residue. For either option there are a wide variety of proprietary systems available, many of which carry guarantees by the supplier. The optimum choice of control system depends upon a wide variety of factors related to the combustor design, fuel feed and air quality standards that must be met. Fluidized bed combustors control emission of sulfur oxides by chemically capturing these pollutants by the bed materials during the combustion process. By controlling the amount of limestone in the bed and the temperature up to 90 percent of the sulfur oxides can be captured. The primary means for controlling oxides of nitrogen is by limiting the combustion temperature to below 1600 degrees Fahrenheit. Fluidized bed combustors are best suited to operate at around these temperatures, while pulverized and stoker units for optimum boiler efficiencies operate at higher temperatures. Comparative Economics Typical capital and annual operating cost for each of the direct combustion systems are summarized in Tables 6.3 and 6.4.2 The cost figures are based on producing steam at 275 psig at 525 degrees F at a rate of 150,000 pounds per hour. Annual capacity 2 Data from Malone, R.A., see references. 125 factor is 70 percent and coal costs are based on $50 per ton for stoker coal, $45 for BFB and $42 for PC and CFB systems. Other operating costs, such as fixed charges for capital, material, power, manpower and waste disposal costs are common to all systems. The capital costs are on an erected basis and include the material handling facilities to the stack. TABLE 6.3 COMPARATIVE CAPITAL COSTS (in thousands 1985 dollars) STOKER PC BFB CFB Boiler and Auxiliaries 3325 5650 6400 6250 Material Handling 1050 1050 1050 1050 Stack 500 500 350 350 F@D System 2000 2000 - - Ash Removal System 430 430 500 500 Baghouse 600 600 600 600 General Construction 1340 1500 1400 1650 Mechanical Construction 800 1000 800 1000 Electrical Construction 900 1050 975 975 I & C Construction 350 450 400 400 Total Capital Cost 11295 14230 12475 12775 Differential Cost base 2935 1180 1480 Data from Malone, see reference. 126 TABLE 6.4 COMPARATIVE OPERATING COSTS (in thousand 1985 dollars) STOKER PC BFB CFB Fixed Capital Charges* 1694 2135 1871 1916 Operating Cost Fuel 2470 1886 2168 1976 Labor 875 875 525 525 Maintenance 113 142 125 128 Auxiliary Power 320 412 463 492 Sorbent ; 71 65 190 123 Waste Disposal 76 69 3il 31 Total Annual Cost 5619 5584 ° 5373 5191 Differential Cost** base (35) (246) (428) * Capital charge rate is 15 percent ** Numbers in parenthesis () are negative values. Data from Malone, see reference. SUMMARY Based on the data in Table 6.3 and the example illustrated, stoker units have an advantage over the other combustion systems. This advantage however, is overcome when the capital costs are annualized and operation and maintenance costs are included. For this example as shown in Table 6.4, the circulating fluidized bed system has the overall lowest cost over the life of the system. 127 The advantages and disadvantages of the four combustion systems compared in this example can be summarized as follows. Stokers Advantages Low capital cost Reliability and high availability Simple operation Reasonable load response No auxiliary fuels required Burns a wide range of coals Low maintenance costs Minimum fouling Low auxiliary power requirements Proven systems Disadvantages Constant operator attendance Requires coal sizing No internal control for sulfur oxides Grate failure shuts down unit Pulverized Coal Systems Advantages Minimum operator attendance High boiler efficiency Excellent load response Burns wide range of coals 128 Minimum preparation of feed coal to pulverizer High availability and reliability Proven systems Disadvantages Requires auxiliary fuels High maintenance on burner components Susceptible to slagging and fouling High auxiliary horsepower required Constant attention to prevent furnace explosions No internal control over sulfur oxide emissions Bubbling Fluidized Bed Advantages Burns a wide range of coals Removes sulfur oxides during combustion Low emissions of oxides of nitrogen Disadvantages Relative poor load response High capital cost High auxiliary horsepower Coal feed preparation required Constant maintenance required Circulating Fluidized Bed Advantages Burns a wide range of coals and other fuels 129 Removes sulfur oxides during combustion Acceptable load response and turndown ratio Low emissions of oxides of nitrogen Disadvantages High Capital cost High auxiliary power requirements Constant maintenance required Technology relatively new CONCLUDING COMMENTS This overview of commercially available, direct combustion systems is intended to show that the advantages, disadvantages and economics will differ for each application. It is important that each user carefully study all possible combustion methods and the potential coal supply options. The final economics can only be determined upon receipt of bids for both fuel and combustion system with appropriate performance guarantees. REFERENCES A Comparison of Circulating Fluid Bed, Bubbling Fluid, Pulverized Coal _an reader Stoker Power Plants, Lutwen, Ronald C., First Annual Fluidized Bed Conference, Council of Industrial Boiler Owners, Washington, D.C., December 1985. nomi f i Pulveri r team Generators, Malone, R.A., First Annual Fluidized Bed Conference, Council of Industrial Boiler Owners, Washington, D.C., December 1985. 130 Stoker Firing STEAM OUTLET \ \ | ' AiR HEATER ‘----- ~---—-----—---] FUEL Ly’ HOPPERS if | | | I \ | TRAVELLING GRATE SPREADER STOKER 131 Pulverized Coal Firing Electrostatic Precipitator Typical pulverized coal fired boiler 132 FLU cre CONVECTION PASS COAL LIMESTONE e& SY FREEBOARD SPLASH ZONE BED de verre a — DISTRIBU" PLATE ~ PLENUM fia RECYCLE FLUIDIZING AIR TRANSPORT AIR cs FORCED ORAFT FAN WASTE WASTE COMPRESSOR SIMPLE AFBC PROCESS SCHEMATIC 133 STEAM OUTLET CYCLINE COMBUSTION CHAMBER © FEEDWATER INLET -— FLUE Gis . LIMESTONE oe me oe REL ——— SECOMNOARY AIR FAN PRIMARY AIR FAN ae? cL TO ASM STORAGE — = CIRCULATING ATMOSPHERIC FLUIDIZED BED 134 Table 1 Simplified Classification of the Various Carbonization Procedures sella eee ee eee EEE eee eee Carbonization Final temp. process °C °F Products Processes eS Low 500-700 930-1290 Reactive coke Rexco (700°C) made temperature and high-tar in cylindrical verti- yield cal retorts. Coalite (650°C) made in ver- tical tubes Medium 700-900 1290-1650 Reactive coke Town gas and gas temperature with high-gas__ coke (obsolete). yield, or Phurnacite, low domestic volatile steam coal, briquettes pitch -bound briquettes carbon- ized at 800°C High 900-1050 1650-1920 Hard, un- Foundry coke (900°C). temperature reactive coal Blast furnace coke for metallur- (950-1050°C) gical use decease EERE Source: Coal and Modern Coal Processing: An Introduction, G. J. Pitt and G. R. Millward (eds.), Academic, New York, 1979, p. 52. Coal Transportation System Earth Fill Removable Bricks for Regulating Air Intake SN TG Ground Surface ARE. AYA ‘Sey FIG. 2 Simplified representation of a beehive coke oven. (From Meyers, 1981.) pr IMPORTANCE OF COKE Source of Heat Reducing Agent Means of Permeability Mechanical Support to Burden CHARACTERISTICS OF COKE Size Control Strength Purity (Chemical Composition) Reactivity DURING CARBONIZATION COKING COALS Soften Agglomerate Resolidify FACTORS INFLUENCING COKE BEHAVIOR Coal Rank Degree of Oxidation Minerals (% Ash, % S) Maceral Composition Coking Conditions Some Comments on the Chemistry of the Agglomerating Tendency of Coals by P.H. Given Fuel Science Section, Material Sciences Department Pennsylvania State University Abstract A simple physical picture of the coking process is presented, in which the production and time of retention of tarry substances inside the pores of coal particles are held to be crucial. The reactions pro- ducing these substances include loss of phenolic OH and hydroaromatic hydrogen. The dependence of the Free Swelling Index on rank and other parameters is discussed, and its value for Predicting the extent of agglomeration in gasification processes is shown to be questionable. It is shown that the agglomerating behavior of coals depends strongly on their petrographic composition as well as rank and rate of heating. Nearly 30 years ago the late Dr. D.H. Bangham, who was then Director of Re- search of the British Coal Utilization Research Association, presented in discussion the following view of the coking process (as far as I know this was never published) . All coals are porous and have a large surface area (100-300 m2/gm) » nearly all of which represents the walls of the very fine pores. When a coal is heated to about 400°C chemical decomposition becomes active » and tarry substances are re- leased into the pores inside the particles with bituminous (coking) coals. These substances cannot readily escape because of the small diameter of the pores and their pore entrances, and so during their residence inside the particles they act as boundary lubricants for colloidal micelles , making the particles behave like a very viscous (non-Newtonian) liquid, so that they agglomerate. With anthracites, insufficient tarry material is formed to act in this way. With lower rank coals, there is a considerably greater proportion of large pores, and so the copious quantities of tarry substances can escape too rapidly to give the par- ticles fluidity. Consequently neither with anthracites nor with subbituminous coals does agglomeration occur. tal proof, nor has any alternative view. Nevertheless it has always seemed to me to be plausible and to provide an easily visualized physical model of real phenomena. It predicts that swelling and caking of bituminous coals should be less marked with very fine particles, which is true (Mackowsky, 1966), and that agglomeration should be increased by rapid heating, which is also true (Mackowsky, 1966). On further heating the tarry substances evaporate and/or decompose, so that the fluid mass re-solidifies. This final phase of the coking process is analogous to the curing of a thermosetting resin such as bakelite or a glyptal. A little is known of the chemical processes occurring in the crucial coking range, 380-500°C, though it is not clear that the knowledge can be usefully ap- plied. Some years ago, J.K. Brown (1955) pyrolyzed a number of vitrains at a series of temperatures at 20° intervals from 360-500°C, and determined the in- frared spectra of the products. The results showed that the first major change in the spectra (occurring at different temperatures according to the rank of the coal) was a decrease in the intensity of the absorption due to aliphatic C-H vibrations. Later work (Peover, 1960; Reggel, Raymond and Wender, 1970) has shown that up to one third of the hydrogen in coals is present on hydroaromatic rings (i.e., rings that by simple loss of hydrogen can become aromatic). Such structures are a special kind of aliphatic material, and are less stable thermally than open chain structures. It is therefore a permissible inference from this and from Brown's findings that the first important chemical change on pyrolysis of bituminous coals is a thermal dehydrogenation or cracking of hydroaromatic structures. Brown also observed that, at a temperature about 20° above that at which aliphatic C-H intensity dropped markedly, the intensity of the pherolic O-H vi- bration decreased sharply, presumably due to dissociations to produce OH radi- cals and hence water. Some curious observations were made some years ago that bears on the caking behavior of coals.* If a bituminous coal is very rapidly heated to about 400°C, held there for a few minutes and then quenched, the product yields about 10% of its weight as a chloroform extract. The extract is extremely soluble in chloroform and if added to a weakly caking coal considerably increases coke quality. The residue after extraction had lost its caking power. Whether this approach might be a viable means of preventing agglomeration of bituminous coals is perhaps. worth consideration. The pyrolysis and extraction steps could possibly be com- bined in one if the "vapor solvent extraction" technique with a solvent above its critical temperature were used. ** Agglomeration is not likely to be a serious problem in liquefaction processes. In processing, some bituminous coals may cake on the walls and escape full ex- posure to reaction conditions. However, in a well-designed continuous flow reactor, such as would be used in any full-scale plant, it should not be difficult * Dryden and Pankhurst, 1955. ** Paul and Wise, 1971. to avoid the phenomenon. It is of course in gasification that the agglomeration of bituminous coals can cause serious problems. How can the probable seriousness of the problem be estimated? The only relevant parameter commonly available is the Free Swelling Index or Crucible Swelling Number. This is determined by heating a small sample in a crucible over a relatively large gas flame, so that heating is rapid. The de- gree of swelling is determined by matching the profile of the coke button produced with one of a set of standard profiles. The result is incorporated as the second digit in the 3-digit code number by which European coals are classified for indus- trial use. Its function is to predict the behavior of coals during rapid heating on a grate, such as a chain grate stoker, before combusticn, as opposed to the func- tion of the third digit, which is to predict behavior during slow heating, in a coke oven. A plot of the Free Swelling Index against organic carbon content (as a rank parameter) is shown in Figure 1 for about 90 American coals. The data are taken from the Penn State Coal Data Base, for which my colleagues William Spackman and Alan David are responsible (and I am indebted to them for the data). At the present time, the Free Swelling Index is available only for a fraction of the coals represented in the Data Base, and there is little reason tc suppose that this frac- tion is a representative sample of U.S. coals. Moreover I arbitrarily selected those coals whose total content of vitrinite + pseudovitrinite exceeded 70% (con- cerning the influence of petrography on agglomeration, see below). It is evident that there is a great deal of scatter in the points and that the line drawn through them is by no means justified. A line has been drawn chiefly to emphasize the flat top to the curve and the very sharp decrease on the high rank side of the maximum, which certainly demonstrate hard fact. It is important to consider why there is so much scatter on the low rank side of the maximum. There are a number of contributory reasons: (1) it is likely that there is a real random element in the dependence of coal properties on rank, (2) the samples represented include bituminous coals from eastern and western States, which have had very different geological histories; it has never been demonstrated whether differing history affects properties of coals at the same level of rank » (3) the samples include coals of differing petrographic composition. In spite of the scatter, it is clear that the Free Swelling Index tends to increase with increasing rank, stays approximately constant for a while, and then sharply decreases. The coking phenomena change considerably with rank, and it is quite likely that different processes are important at different levels of rank. It is the bituminous coals of intermediate rank (high volatile A bituminous) that show the greatest fluidity in the range 400-500°C, so that with them agglomeration in a flu- idized bed would resemble the condensation of a cloud of liquid droplets. At the lower and upper ends of the rank range, agglomeration would be more like the 9.5 a 7 ] 1-113 12 a a | - WN1 1 8.0 1 11 1 7 1 6.5+ 1 a 7 © 7 a} £ 5.0+ 1 1 rl 7 £ 7 1 1 2 3 - 1 é - i 3.5+ 1 1 1 $ 7 rr - 1 | 7 al 1 2.0+ 1 1106¢42~«0«2 - 1 - 1 05+ 1 4131 7°21 $enn------ $onneoee-- teeccennee $eo------- t--------- $--------- too --- eee $occeee 77.5 82.5 87.5 92.5 2 80.0 85.0 90.0 95.0 i Mie %C, d.mmé Approximate ASTM — sbb and HVC HVB HVA | MV LV Anthracite rank class Figure 1 agglomerating tendencies of bituminous coals. Coals are heterogeneous, not only in a chemical sense; they contain a number of components of different origin, which are also of radically different behavior on Pyrolysis. The principal component (or maceral) in most coals is vitrinite, derived from the coalification of wood. It is vitrinite that has the coking properties discussed above. It is often accompanied by lesser amounts of a closely related maceral, Pseudovitrinite, which is usually inert in the coking process in the sense that it does not swell or become fluid on heating, and does not contribute to agglomera- tion. Other macerals, fusinite and micrinite » are inert in the same sense, and moreover yield little volatile matter. The other maceral commonly found in coals of the eastern U.S. (found to a lesser extent in western coals) is sporinite (or exinite as it is sometimes called). Sporinite is usually present in amounts of about 3-8%, but its properties are so different from those of the other macerals that even at this level of concentration it may affect the agglomeration of a coal. The volatile matter yield can be 60-80%, much of which is lost at temperatures below 500°C. At about 400° it becomes ex- tremely fluid and when the pure maceral is heated it blows up into a thin walled bubble structure; when this resolidifies it has almost no mechanical strength. In The average vitrinite content in most coal seams is 60-70%. Since the proper- ties of the various macerals differ so widely, it is clear that the agglomerating tendencies of coals will vary considerably according to the distribution of macerals in the 30-403 that is not vitrinite. This is Probably the chief reason for the large degree of scatter in Figure 1. The steel industry has developed empirical formulae by which coke strength and reactivity can be successfully predicted for single coals or blends from the petrographic analyses. No doubt similar formulae could be developed for agglom- erating tendency in gasifiers, once a good quantitative measure of this tendency has been established. There are a few coal seams containing very large reserves that contain more complex mixtures of macerals and less vitrinite than usual. An example is the Elkhorn No. 3 seam in Kentucky. It is possible to modify a coal preparation plant to yield product streams of differing petrographic composition. A plant could be designed to obtain from the Elkhorn coal one stream rich in vitrinite (say 10-20% of the whole), and one poor in vitrinite. The latter should gasify well without agglomeration, even though the rank of the vitrinite is HVA (I am indebted for this suggestion, which has not yet been tested experimentally , to Dr. William Spackman). References 1. J.K. Brown (1955), Infra-red spectra of coals, J. Chem. Soc., 744; Infra- red spectra of carbonized coals, ibid., 753. 2. 1I.G.C. Dryden and K.S. Pankhurst (1955), Plastic softening of coking coals on heating, Fuel, 34, 363. 3. M.-Th. Mackowsky and E.-M. Wolff (1966), Microscopic investigations of pore formation during coking, "Coal Science ," Vol. 55 in Advances in Chemistry Series, Amer. Chem. Soc., p. 527. 4. M.E. Peover (1960), Dehydrogenation of coals with benzoquinone, J. Chem. Soc., 5020. 5. P.F.M. Paul and W.S. Wise (1971), "Principles of Gas Extraction ," Monograph CE/5, Mills and Boon, London. 6. L. Reffel, 1. Wender and R. Raymond (1970), Catalytic dehydrogenation of coal. Part 4. A comparison of exinites, micrinites and fusinites with vitrinites, Fuel, 49, 281, and earlier papers referenced therein. U. S. Steel Corporation Research Laboratory Monroeville, Pa. 15146 "SELECTION OF COALS FOR COKE MAKING" By R. J. Gray, J. S. Goscinski and R. W. Shoenberger (Presented by J. W. Robinson at the conference sponsored jointly by the Iron and Steel Society of AIME and the Society of Mining Engineers (SME) of AIME on October 3, 1978 at the Greater Pittsburgh Airport Holiday Inn, Pittsburgh, Pa.) SELECTION OF COALS FOR COKE MAKING R. J. Gray, J. S. Goscinski and R. W. Shoenberger Abstract This report has been prepared to describe coal properties and their relative importance in formulating blends for producing high-quality metallurgical coke. To qualify as coking coal, the coal must be classified bituminous and subclassified between high-volatile B and low volatile in rank. It must also be agglomerating and have the capability to melt when heated. For prime quality, the coal should be strongly coking and have a minimum amount of noncarbon impurities such as silica, alumina, iron, calcium, sulfur, phosphorus, chlorine, sodium, and potassium. To obtain high- strength coke, the individual coals must be blended to produce blends that range in volatile-matter content between 25 and 32 percent (daf) and in vitrinoid reflectance between 1.1 and 1.3 percent. Although coke Cech improves as the rank of the blend increases (as volatile matter decreases and reflectance increases), the amount of improvement is controlled by the inert-maceral content of the blend and is limited by carboniza- tion pressure and contraction obtained during coking. Few coals have all the properties desirable for coke making. However, a deficiency in one property of a coal in a blend can be offset by an excess of that property in another -la- coal, and as a result, trade-offs can be made in coal selection to formulate satisfactory blends. In addition, the strength of coke from a particular coal blend can be improved by coal- Preparation and coke-oven-operating practices, such as pulverizing the coal to a smaller particle size, decreasing the coking rate, or increasing the coal-charge bulk density. -2- Introduction Coal is an energy resource composed of a mixture of organically derived macerals and associated minerals. Macerals are the plant remains that have undergone chemical and physical changes in response to geologic processes. The kinds and amount of each maceral present determine the coal type. The degree of metamorphism or alteration of the macerals establishes the coal rank, and the amount and type of minerals associated with the organic constituent determine the coal grade. Industrial coal petrography deals with the microscopic determination of how coals differ in type, rank, and grade and how these differences affect the utilization of coal. Labora- tory carbonization testing deals with quantitatively measuring the effect of rank, type, and grade on expansion/contraction, pressure, and coke strength. In this report, the ranges for the coal properties that determine coking characteristics are presented for use as a guide in assessing the quality and utilization of metallurgical coals. In addition, their relative importance in establishing the criteria for selecting coals in the formulation of blends for the production of high- quality metallurgical coke is discussed. Blast-Furnace-Coke Specifications The important coke properties that affect blast- furnace performance are chemical composition, size, strength, and reactivity. Coke composition is generally measured in terms Of volatile-matter, ash, sulfur, alkali, and phosphorus contents. A good coke has a volatile-matter content of less than 1 percent, an ash content of 9.0 percent or less, and a sulfur content of 0.8 percent or less. The optimum size of blast-furnace coke should be 3 inches by 1 inch (76 by 25 mm). Coke strength is measured by determining the amount )* of degradation induced by shatter (ASTM p3038) + 2) or by tumbling (ASTM D3402). The indexes of coke strength from the tumbler test are called the stability and hardness factors. The stability factor is the most commonly used strength criterion employed in the United States and thaventeg the tendency of the coke to break upon handling and impact. The hardness factor indicates the tendency of the coke to abrade into fines upon handling. Other strength tests (Micum, Irsid, JIS, Sundgren) used throughout the world have been related to the ASTM tumbler test. In general, blast-furnace performance improves with increased coke stability. This is particularly true for large-diameter blast furnaces. Coke reactivity tests indicate the rate at which carbon is converted to carbon monoxide by reaction with carbon dioxide under specified conditions of temperature and gas * See References. -4- flow. Generally, the higher the rank of the coal blend and the higher the final coking temperature, the lower the reactivity of the coke. Few coals meet all the requirements to produce a high-quality coke, so coals must be blended to meet these requirements. Thus, to evaluate a specific coal for coke Making, it is important to know what other coals are to be used in the blend and how much of each is to be used. Frequently, in commercial blends it is economically attractive to include coals that do not meet normal metallurgical-coal specifications. In these cases, it is necessary that the other coals in the blend be correspondingly higher in quality or have properties that compensate for deficiencies in the poorer coal. Coal Properties Coal Rank All coking coals posses the unique property of softening, agglomerating or fusing, and resolidifying to form a coherent, porous coke structure during carbonization. The class of coals referred to as bituminous are the only coals in the lignite-to-anthracite rank series that possess these Properties. In addition, it is only the agglomerating bituminous Coals that are considered coking and caking. Coals are classified (ASTM D388) according to rank, and rank is the most important parameter relating to the -5- coking potential of coals. Figure 1 compares the rank group, names, and boundary lines of the ASTM system with class names and boundary lines of the international system, >) Bituminous coals are subdivided by rank into high-, medium-, and low- volatile coals, with the high-volatile bituminous coals sub- divided into high-volatile A, B, and C. In general, high- volatile C coals are noncoking, high-volatile B coals are Marginal coking, and high-volatile A coals are coking. The medium-volatile bituminous coals and some of the high-volatile A-rank coals that are nearer the medium-volatile coals in rank are good coking coals and, like medium-volatile coals, can be used individually to make strong coke. Although low-volatile and high-rank medium-volatile coals produce strong cokes, they exert excessive wall pressure during carbonization and cannot be coked alone in by-product ovens since they can cause oven damage. In addition, these coals do not contract sufficiently during coking to permit easy removal from the coke ovens. Table I gives a range of coking properties of the ranks of coals (high, medium, and low volatile) that are used to produce a coking-coal blend with an acceptable coke strength. These coals are classified by rank according to dry, mineral- Matter-free volatile matter (ASTM p388).*) Although volatile Matter (Property 1, Table I) is a convenient and universally accepted indicator of coal rank, it is one of the least reproducible chemical properties determined on coal and is affected by the presence of carbonates and changes in petro- graphic composition. The rank of a coal is also determined in petrographic analysis” by measuring the mean maximum reflectance of the vitrinoid maceral, the major reactive component in coal (Property 2, Table I). The relationship between these two rank parameters, vitrinoid reflectance and volatile Matter, is 6) Since the reflectance of vitrinoids shown in Figure 2. is not influenced by other coal properties as volatile matter is, reflectance is a more accurate measure of the relative rank difference between coals. Reflectance measurements on individual vitrinoid macerals in bituminous coal may range from about 0.5 to 2.00 percent reflectance. Vitrinoid-reflectance measurements are classified into types, each type representing a reflectance range of 0.10 percent. For example, V8 contains all the vitrinoids with a reflectance of 0.80 through 0.89 percent. Coal Type In addition to rank, coking coals are also selected by type or the petrographic reactive- and inert-maceral content. The maceral composition is determined microscopically (ASTM 7) D2799). Coals are commonly classified in terms of bright or dull types. Bright coals are generally considered superior to dull coals for coke making. Bright-banded bituminous coals macroscopically consist of an abundance of clarain and vitrain, and microscopically the clarain and vitrain contain a predominance . of the reactive maceral vitrinoid and may have significant amounts of resinoid and exinoid macerals. The vitrinoid macerals soften and resolidify to form the continuous-bond phase during carbonization. The exinoids and resinoids Produce mostly by-products but also contribute to the bond phase in coke. Dull-banded bituminous coals Macroscapically consist of an abundance of durain, and microscopically the durain commonly contains a greater abundance of the inert macerals. The inert macerals do not soften during carbonization and act as inert filler in the coke structure. Dull coals can be used to a limited extent in blends, particularly when the dull coals have good chemistry or price or some other advantage that would give an incentive for including them in blends. The organic inert macerals are composed of micrinoids, fusinoids, and semifusinoids. In North American coals, 2/3 of the semifusinoids are categorized as inert macerals in high- volatile coal and 4/5 are classed as inert macerals in low- volatile coal. The organic inerts will burn in the blast furnace. Their inorganic inerts are the ash-forming materials or mineral matter composed largely of silicon, aluminum, iron, calcium, and alkalies such as sodium and potassium. The inorganic inerts will not burn in the blast furnace. It has been found by laboratory experimentation that each rank of coal (measured by the reflectance of the vitrinite) has an optimum inert content for best coke strength. The relation of vitrinoid type to the optimum inert content is 8) The ratio of the actual inert content of shown in Figure 3. a coal to its optimum inert content is called the composition- balance index. Thus, by definition, a composition-balance index of 1.0 would give the optimum coke strength for that coal. Either an excess or a deficiency of inerts would result in decreased strength of coke from a coal of a given rank. However, in the case of coal blends, a deficiency of inerts in one coal can be at least partly offset by a surplus of inerts in another coal, assuming high pulverization levels and proper proportioning and mixing. The rank of a coal affects to some degree the amount of inerts that result in optimum coke strength. As the coal rank increases from high volatile to medium and low volatile, the optimum inert content decreases because the vitrinite (principal coke-producing maceral) of the higher rank coals cannot assimilate the inerts as well as the lower rank fluid- type vitrinite. The optimum amount of inerts for most coking- coal blends with a vitrinoid reflectance of 1,3 percent is about 15 percent, whereas the optimum amount of inerts for blends with a vitrinoid reflectance of 1.2 percent is about -9- 25 percent. In general, blends with lower reflectance and higher Gieseler fluidity can tolerate and incorporate more inert materials. Coal Fluidity Various tests have been proposed to measure the ability of coking coals to incorporate inerts. Two of the most commonly used tests are the Gray King Test (150/R505) 2? and the Roga Test (United Nations Publ. 1956 11.E.4, E/ECE/247, E/ECE/Coa1/110) .+) In the United States, the Gieseler plasto- ) meter test (ASTM p2639) + is commonly used to measure the plastic properties of coal during heating (Property 3, Table I). Coking coals soften, then become very fluid, and finally solidify. Strongly coking high- and medium-volatile coals become very fluid and have a wide fluid range. The fluid range is defined as the difference between the solidification and softening temperature. Poor-coking high- and low-volatile coals have low fluidity and a narrow plastic range. In selecting coking coals for blends, the coals should have widely overlapping plastic ranges to assure the 2) in production of homogeneous coke structure, Figure 4.2 addition, many workers in the field insist that the blend fluidity should exceed 2000 dial divisions per minute. In addition, some workers believe the coke contraction relates to coal fluidity in addition to rank. The Arnu or Ruhr dilatometer is used widely in Europe to measure plastic properties of -10- 15) coals (IS0/TC/27, Doc. 221,+3744) AFNOR and DIN 51739, respectively). These tests are becoming more popular in the United States. The free-swelling index (FSI) test (ASTM p720)+®) is used to determine the agglomerating or swelling properties of coal (Property 4, Table I). In this test, the coke-button height is used to judge the caking and swelling properties of acoal. In general, the FSI of coking coals should exceed a coke-button size of 4; however, the better coking coals have a button size in excess of 7. Australian workers have shown that changes in the coke-button size are a function of rank and the proportion of vitrite and clarite present ina coal.?”? Hardgrove Grindability Index The Hardgrove grindability index (HGI) (ASTM p409-71) +8? measures the hardness, strength, and fracture characteristics of coal. It is used to determine the relative gz indabi lity or ease of pulverization of coals in comparison with coals chosen as standards. In this method, a prepared sample of known size-consist receives a definite amount of grinding energy in a miniature pulverizer and the change in size-consist is determined by sieving. The higher the index, the easier the coal is to pulverize. The ease or difficulty of pulverizing coal is mostly _ a function of the coal rank (Property 5, Table I). High- volatile coals are difficult to pulverize and the indexes vary -ll- between about 32 and 75. The lower rank high-volatile coals have indexes between about 32 and 70 and the higher rank coals, between 48 and 75. The lower rank medium-volatile coals have indexes between about 60 and 90 and the higher rank coals, between about 80 and 135. The lower rank low-volatile coals have indexes between about 90 and 120 and the higher rank coals, between about 85 and 105. At the upper limit of medium-volatile coking-coal rank, the indexes start to decrease until low values are obtained on noncoking anthracites (20 to 45 range). Other properties of coal such as type of ash and petrographic-maceral content also affect the grindability index, although to a lesser extent than rank. The relation of HGI to coal rank (volatile matter) and type (maceral content) is shown in Figure 5.19) A higher ash content can increase or decrease the HGI, depending on the rank of coal and type of ash; a higher micrinoid, exinoid, and/or resinoid content will reduce the HGI. The breakage of coal and the ease or difficulty of pulverizing coal is of considerable importance in coal-washing plants and in the preparation of coal for charging to the coke ovens. Coke plants utilize high-volatile, low-volatile, and/or medium-volatile coals in the coal blends to obtain acceptable coke strength. Because of the great difference in -12- the HGI, it is most desirable to pulverize each coal rank and type separately. One scheme has been proposed in which HGI and volatile matter of the coal are used to predict coke stability. 7?) Coal Grade Coking coals are also selected by grade, in addition to rank and type (Table II). Coal grade principally relates to the chemistry such as ash, sulfur, alkali, chloride, and Phosphorus contents. In addition, the chemistry of the ash and the ash-fusion characteristics are often determined. In general, the ash content should not exceed 8 percent, and the better coals have an ash content of 6 percent or less. The sulfur content should not exceed about 1 percent, and the better coals should have a sulfur content of less than 0.7 per- cent. The phosphorus limitations relate to the other burden materials and, for normal blast-furnace operation, the phosphorus in the hot metal should not exceed 0.1 percent. U. S. Steel's limits for various chemical components of blast-furnace burdens are shown in Table 111.22) If any of the chemical components of a prospective material exceed these limits, the material may still be satis- factory for specific applications because it might be possible to blend it with other materials to keep the composition of the entire burden within the specified limits. -13- The limits for the components in Group I are based on the chemical specifications for residual elements for many grades of steel. In plants where these steels are not produced, some exceptions to these limits would be permitted. The limits for the components in Group II are based on the maximum amounts that can be tolerated without encountering severe operating difficulties or serious environmental pollution problems. The limits for the components in Group III are based on amounts in excess of which higher than normal operating cost would be encountered because of additional fuel requirements or refining times. No limit is set for manganese because the requirements for manganese differ greatly from plant to plant. In general, all the components in Group III should be as low as possible. The alkalies attack and break the coke, and also cause scabs and other operating problems in the blast furnace. Therefore, the alkali content should be kept as low as possible. Chlorides pass into’ the coke-plant by-product system and require considerable water to remove them from the tars. In addition, chlorides cause maintenance problems both in the Coal-preparation plants and coke works because of the corrosive nature of the compounds containing chlorine. -14- The ash-softening temperature for coking coals should be relatively high (+2300°F) so -that the coal ash does not fuse to the refractory coke-oven lining during carbonization. Although coal grade is an inherent property of the coal, the grade can be and is established by exploration property evalua- tions and improved to various degrees by beneficiation plants. Coal Oxidation In addition to coal rank, type, and grade, particular emphasis is also placed on detecting oxidized coal since it can adversely affect coal-charge bulk-density control, coal 22) The flow, coke strength, and coking characteristics. extent of coal oxidation can be measured by heating a minus- 100-mesh sample in a caustic solution (NaOH) and recording the light transmission of the filtered solute with a spectrophotometer. Since oxidized coal is soluble in caustic, the light transmit- tance of a coal is lowered when oxidized coal is present. Certain blends containing coals with less than 80 percent transmittance have proved difficult to handle, and control of bulk density has been a problem. In addition, some success has been achieved in correlating the percent and degree of oxidized particles in different ranks and types of coal, as determined microscopically, with their light-transmittance values. A test for detecting oxidized coal is outlined in Table Iv. -15- Calculated Coke-Stability Factor Based on data obtained from more than 300 laboratory carbonization tests on all ranks of coals and coal blends, an excellent correlation was established between the coke-stability factors determined on coke produced in laboratory carbonization tests and those calculated from petrographic composition and reflectance analyses on the individual coals used in testing. These tests were conducted with standard conditions of pulveri- zation of the coals to 80 percent minus 1/8 inch (3.2 mm), a coking rate of 1.03 inches (26.2 mm) per hour, and a bulk density of about 53.5 pounds per cubic foot (857 kg/m?) , Two indexes are required to calculate the coke- stability factor of a coal. First, the composition-balance index which, as discussed earlier, is the ratio of the actual determined total inert content of the coal to the optimum inert content for best coke strength for the particular rank of the coal as determined by reflectance (Property 6, Table I). Second, the rank index, which represents the relative effect of coal rank on coke strength as measured on a scale from 2 to 8 (Property 7, Table I). The rank index is determined by proportionally combining the coking strength of the individual vitrinoids at a given inert level that make up the coal. The relation between vitrinoid reflectance types is plotted with reference to inert content and rank index on -16- Figure 6. In general, the rank index increases as the reflectance of the vitrinoids increases up to 1.99 percent (Vitrinoid 19), after which the rank index decreases. However, for any given vitrinoid, the rank index is highest at the optimum inert level, and decreases with an excess or deficiency of inerts. The composition balance index and rank index are used to predict the calculated coke-stability factor from a graph, (Property l, Table V), Figure 7. The isostability curves are curves of equal stability and are based on the laboratory coke tests on blends and individual coals.°’23) To obtain the coke-stability factor expected from a given coal or coal blend carbonized under plant operating conditions, corrections must be made, taking into account those operating factors that were different from the standard conditions for which the original correlations were obtained; that is, pulverization level, coking rate, and charge bulk density. Because the effects of these three operating variables on coke stability are not the same for all coal blends, labora- tory carbonization tests must be conducted to establish general relationships for adjusting to plant conditions. Generally, coke strength is increased as the pulveriza- tion level and bulk density are increased and as coking rate is decreased (Figures 8, 9, 10). In addition, the lower the rank of the blend the larger the increase in coke strength for each of these operating variables. -17- Coal Blends Only about 10 percent of the coking-coal reserves are of medium-volatile rank, 8 percent are low-volatile, and the remaining 82 percent are high volatile, *4) Because there are insufficient reserves of medium-volatile coals to permit their exclusive use in coke making, and because these coals exert high pressure and have low contraction and cannot be coked rm slot-type ovens, industry has resorted to blending of high-, medium-, and low-volatile coals. The blends commonly consist of 60 to 85 percent high-volatile coal with 15 to 40 percent low- and/or medium-volatile coals. The rank of the coal blend is generally controlled to a volatile matter of 25 to 32 percent (daf), which corresponds to a vitrinoid reflectance of 1.1 to 1.3 percent. Some experts prefer a blend reflectance between 1.1 and 1.2 percent. When a blend is on the lower volatile end of the range (higher reflectance), it is necessary to operate at lower bulk densities and possibly lower coking rates than when the blend contains higher volatile matter. This precaution is taken to avoid high pressure and insufficient contraction during carbonization. The relation of reflectance of vitrinoids to coke stability, expansion/contraction, and pressure for individual coals is shown in Figure ll. -18- Volume Change During Carbonization To assure easy pushing of the coke from the ovens at the end of the coking cycle, the coke must contract away from the oven walls. The volume-change characteristics of coals and coal blends are determined quantitatively in the sole-heated oven (ASTM D2014) .*>) In this method, a known weight and thickness of coal is heated from the bottom surface of the charge while a specified force is applied to the top by a piston. At the end of the test, the thickness of the coke is measured by recording the final position of the piston, Experience has shown that the rank of Lhe coal blend, coal-charge bulk density, plastic properties, and total inert content of the coal charge control the volume-change characteristics of coals. Table v, Property 2, shows the expansion/contraction characteristics of individual high-A, medium-, and low-volatile rank coals. High-volatile coking coals contract significantly when carbonized alone. Medium- volatile coals at the low end of the rank scale contract sufficiently, but at the upper end of the scale they exhibit expansion and normally cannot be used alone to produce coke. Low-volatile coals are normally expanding and cannot be used alone to produce coke because they cannot be pushed from the ovens. The general relation of volume change to individual -19- coal reflectance (rank) is shown in Figure 127°) and the general relation of blend-coal reflectance to volume change is shown in Figure 13. To make acceptable-strength (stability) coke, medium- and/or low-volatile coals that can be used are limited by both the expansion/contraction and the coking- pressure properties of the blend. In addition to the coals used in a blend, the bulk density of the coal charge in the oven has a significant effect on the expansion/contraction properties of coals during coking. As bulk density is increased, contracting coals become less contracting and expanding coals become more expanding. Therefore, a slightly expanding coal ata higher bulk density Can be made to contract within limits by reducing bulk density. The expansion/contraction properties of coals are also affected by their fluidity and total inert content. With coals of a given rank (reflectance), those with higher inert content or lower fluidity will contract less or expand less than coals with lower inerts or higher fluidity in somewhat the same manner. In establishing the required contraction of coal blends for the various coke plants, several factors have to be considered. Among these factors are (1) the condition of the coke ovens and the amount of carbon on the walls, (2) the amount of oven taper from pusher side to coke side, (3) the -20- coal segregation experienced during handling and charging the oven, (4) the capability of the facilities to accurately proportion the various coals used in the blends, (5) the Capability of the facilities to control bulk density by either oil or water addition to the coal charge, and (6) the variability in rank of the coals used, especially the low-volatile coals. Where most of the above factors are favorable, a smaller safety factor is used. In plants where most of the factors are unfavorable, larger safety factors are used which result in the necessity to use lower coal bulk densities with a resulting loss of coke production. The actual contraction required for each plant is based mostly on past experience with hard pushes and stickers. Most plants require between 5 and 12 percent contraction of the charge for easy pushing of the coke, Coking Pressure Coking pressure results from the gas pressure developed in the coal plastic layer and on the coal side of the plastic layer during carbonization. This gas pressure is related to the permeability of the plastic layer and the evolution of gases. The pressure is exerted from the plastic layer through the coke to the oven walls. Coking pressures are determined in 30- or 500-pound experimental Pilot-scale test ovens con- taining a movable wall on which the total force is measured. -2l- These coking pressures have been related to commercial ovens through the gas pressures developed in both types of ovens. After the coal has been charged to the ovens, the heat front generated by the walls on each side of the ovens at temperatures of about 2400°F moves into the coal mass which starts to become plastic. At the same time, gases are being driven out of the coal between the plastic fronts coming from both sides of the oven. The gases are partly prevented from escaping through the plastic fronts, and pressure starts to build up in the envelope between the two plastic fronts, exerting pressure on the oven walls. A peak pressure occurs as the two plastic fronts meet, which occurs after about 12 hours through an 18-hour coking time. The pressure then decreases because the plastic mass has solidified, permitting the gases to escape through the cracks and fissures. A peak pressure does not occur if a plastic envelope does not form or if the plastic layer is very permeable. The rank of the coal primarily determines the coking- pressure characteristics of the coal, as shown in Table v, Property 3. The bulk density of the coal charge has a significant effect on the coking pressure because, as the bulk density increases there is more coal per cubic foot of oven volume, which means the coal is packed tighter in the oven. The -22- Plastic mass is then less permeable to gas flow, and the pressure builds up to a greater degree than at lower bulk densities. Figure 14 shows the general relation of bulk density to pressure. Total inert content of the coal has an effect on pressure. The general relation of pressure to vitrinoid reflectance (rank) and inert level of the coal is shown in 7) and 16.78) Figures 1s? With coal of a given rank, the pressure exerted is less with a high inert content than with a low inert content at the same coal bulk density. This can be explained by the fact that the inerts do not become plastic and an increase in inerts simply reduces the amount of material in the coal that becomes plastic during coking. In addition, the inert content affects the coal plastic Properties which have a great influence on coking pressure, The coking rate used with a given coal blend will affect the coking pressure; however, various coals and blends respond differently and the relationship between coking rate and coking pressure must be determined from laboratory carboniza- tion tests. Most coke ovens are designed to withstand lateral pressure of 2 pound per square inch (psi) (14.06 kN/m?) or more. To assure that ovens are not damaged by excessive pressure, the same factors listed for contraction of coal -23- blends must be considered. Most coal blends are designed to exert no more than about 0.5 to 1.5 psi (3.515 to 10.55 kN/m?) at the normal operating conditions. Conclusions In conclusion, to qualify as coking coal, the coal Must be classified bituminous and subclassified between high- volatile B and low volatile in rank. It must also be agglom- erating, and have the capability of melting when heated. For prime quality, the coal should have a minimum amount of noncarbon impurities such as silica, alumina, iron, calcium, sulfur, phosphorus, chlorine, sodium, and potassium. To obtain high- strength coke, the individual coals must be blended to produce blends that range in volatile-matter content between 25 and 32 percent (daf) and in vitrinoid reflectance between 1.1 and 1.3 percent. Although coke strength improves as the rank of the blend increases (as volatile matter decreases and reflectance increases), the amount of improvement is affected by the inert-maceral content of the blend and is limited by carboniza- tion pressure and contraction obtained during coking. Few coals have all the properties desirable for coke making. However, a deficiency in one property of a coal ina blend can be offset by an excess of that Property in another ' coal, and as a result, trade-offs can be made in coal selection to formulate satisfactory blends. In addition, the strength -24- of coke from a particular coal blend can be improved by coal preparation and coke-oven operating practices such as pulverizing the coal to a smaller particle size, decreasing the coking rate, Or increasing the coal-charge bulk density. 10. 11. 12, =25-= References D3038-72, 1977, 1977 Annual Book of ASTM Standards: Gaseous Fuels; Coal and Coke; Atmospheric Analysis, 01-026077-13, Part 26, Standard Method of Drop Shatter Test for Coke, pp. 363-368. D3402-76, 1977, (see Reference 1), Standard Method of Tumbler Test for Coke, pp. 417-419. Montgomery, W. J., 1976, “Coal Classification, National and International Standards," Symposium on Coal Evaluation, October 31-November 1, 1974, Calgary, Alberta, Information Series 76, p. 15. D388-77, 1977, (see Reference 1), Standard Specification for Classification of Coals by Rank, pp. 214-218. Schapiro, N., and Gray, R. J., 1970, "Petrographic Classifica- tion Applicable to Coals of All Ranks," Proceedings of the Illinois Mining Institute, 68th year, pp. 83-97. McCartney, J. T., 1970, Report on Studies of Rank Classifica- tion of Coal by Reflectance for a Task Group of Subcommittee XVIII, Committee D-5, ASTM (unpub), 7p. D2799-72, 1977, (see Reference 1), Standard Method for Microscopical Determination of Volume Percent of Physical Components of Coals, pp. 359-362. Schapiro, N., Gray, R. J., and Eusner, G. R., 1961, "Recent Developments in Coal Petrography," Proceedings of the Blast Furnace Coke Oven, and Raw Materials Committee, Vol. 20, Pp. 89-112. Williamson, I. A., 1967, Coal Mining Geology, London, Oxford University Press, p. 229. Van Krevelen, D. W., and Schuyer, J., 1957, Coal Science, Amsterdam, Elsevier Publishing Company, pp. 16,19,24,25. DB2639-74, 1977, (see Reference 1), Standard Test Method for Plastic Properties of Coal by the Constant-Torque Gieseler Plastometer, pp. 327-333. Szadlow, A. J., 1976, “Laboratory Testing of Metallurgical Coals," Symposium on Coal Evaluation, October 31-November ae 1974, Calgary, Alberta, Information Series 76, p. 43. 13, 14, 15. 16. 17. 18. 19. 20. 21. 22. 23. -26- Loison, R., Peytavy, A., Boyer, A. F., and Grillot, R., 1963, "The Plastic Properties of Coal," in Lowrey, H.H., ed., Chemistry of Coal Utilization, New York, John Wiley and Sons, Inc., pp. 153-156, Staff, Office of the Director of Research, 1967, "Method of Analyzing and Testing Coal and Coke," U. S. Bureau of Mines, Bull. 638, pp. 43-46. Walters, J. G., Ortuglio, C., and Wolfson, D. E., 1971, "Prediction of Coke Strength of American Coals by the Ruhr Dilatometer Method," U. S. Bureau of Mines Report of Investi- gations 7564, pp. 1-18. D720-67, 1977, (see Reference 1), Standard Method for Free- Swelling Index of Coal, pp. 244-250. Brown, et al., in Stach, E., et al., ed., Stach's Textbook of Coal Petrology, Stuttgart, Gebruder Boratraeger Berlin, Murchison, D. G., trans., pw i351, D409-71, 1977, (see Reference 1), Standard Test Method for Grindability of Coal by the Hardgrove Machine Method, pp. 220-226. : Hsieh, Shuang-shii, 1976, "Effects of Bulk Components on the Grindability of Coals (abs) ," unpublished PhD Thesis, the Pennsylvania State University. Leonard, J. W., 1973, "Evaluating Coking Coals," 1973 Keystone Coal Industry Manual, New York, McGraw-Hill Inc., Mining Informational Services, pp. 394-396, Faigen, M. R., Rygiel, R. J., and Stephenson, R. L., 1972, "Limits for Various Chemical Components in Blast Furnace Burdens," U. S. Steel Internal Memo, January 1972. Gray, R. J., Rhoades, A. H., and King, D. T., 1976, “Detection of Oxidized Coal and the Effect of Oxidation on the Technological Properties," Transactions of SME (AIME), Vol. 260, 8p. Schapiro, N., and Gray, R. J., 1964, "The Use of Coal Petrography in Coke-Making," Journal of the Institute of Fuels, Vol. XI, No. 30, 9p. 24. te ou . 26. 27:21 | 28, -27- Sheridan, E. T., 1967, “Source and Magnitude of Coking Coal Reserves in the United States," presentation before American Coke and Coal Chemicals Institute, Western Regional Meeting, February 1, 1967, Chicago, lllinois, 22p. b2014-71, 1977, (see Reference 1), Standard Test Method for Expansion or Contraction of Coal by the Sole-Heated Oven, pp. 287-292. Splitstone, D. E., 1970, "The Use of Coal Petrography for Predicting the Expansion Characteristics of Coking Coals and Coal Blends," U. S. Steel Internal Memo, July 1970. Gray, R. J., and Waslo, S., 1971, "Predicting Peak Coking Pressure from Coal Petrography," U. S. Steel Internal Technical Report. Thompson, R. R., Shigo, J. J., III, Benedict, L. G., and Aikman, R. P., 1965, "The Use of Coal Petrography at Bethlehem Steel Corporation," Joint Meeting of the Eastern and Western States Blast Furnace and Coke Oven Association, Pittsburgh, Pa., November 11-12, 1965. Table I Rating of Coking Coale for Blending Coal Classification High Volatile-A Medium Volatile* Low Volatile Rating _ Rating Rating Property Good Medium Poor Good Medium Poor Good Medium Poor Volatile Matter, % 31.0-33.0 33.0-36.0 +36.0 21.0-24.0 24.0-27.0 27.0-31.0 18.0-21.0 15.0-18.0 <15.0 Vitrinoid Reflectance, I** 0,92-1.09 0.85-0.95 0.68-0.85 1.40-1.50 1.20-1.40 1.10-1.20 1.51-1.70 1.70-1.85 21.85 Fluidity, ddpm*** +20,000 5000-20, 000 <5000 500-8000 300-20,000 <300-—>20,000 100-300 30-1000 <30->1000 Free-Swelling Index 9 6-8 <6 9 7-8 <7 9 7-8 <a Hardgrove Grindability Index 48 -75 32-70 80-135 60-90 90-120 85-105 Composition-Balance Index** 0,40-0.80 0,.80-1.40 >1.4 1.0-1.50 1.50-2.00 >2.0 2.00-3.50 3.50-5.00 >5.00 Rank Index** 3.4-4.3 3.0-3.4 2,2-3.0 6.0-6.5 4.3-5.5 <4.3 >6.8 6.0-7.5 <7.5 ' coals * Those properties such as volatile-matter content, reflectance in oil, and rank index have little bearing in the ranking of mealum-volatile because the rank required for a medium-volatile coal is dependent upon the rank and amount of the other coals used in the blend. ** Determined petrographically *** Dial divisions per minute (01-D-515-007-7) Ash, % Sulfur, $% Table II Chemistry of Coking-Coal Blends Potassium and Sodium Oxides, % of ash Ash-Fusion Temperature, °F Phosphorus*, % (01-D-515-007-7) Good <6.0 <0.7 >2500 <0.01 Acceptable Table III Proposed Limits for Various Chemical Components of Blast-Furnace Burdens Including Ore, Stone, and Coke Maximum Amount of Maximum Ratio of Component Expressed Component Component to Fe as 1b per ton Fe Group I Cu 0.0001 0.2 Ni + Co - 0.0004 0.8 Mo 0.0003 0.6 Sn 0.00015 0.3 cr 0.0004 0.8 Vv 0.0001 0.2 Group II Zn 0.0004 0.8 Pb 0.0008 1.6 TiO, 0.01 20.0 Na20 + K50 0.002 4.0 As 0.0001 0.2 Sb 0.0001 0.2 P 0.002 4.0 s 0.001 2.0 cl 0.0001 0.2 F 0.0002 0.4 Group III Sid02 0.1 200.0 Al203 _ 0.04 80.0 Mn (Each application must be considered separately} Group I - Limits for the components are based on the chemical specifications for residual elements for many grades of steel. Group II - Limits for the components are based on the maximum amounts that can be tolerated without encountering severe operating difficulties or serious environ- mental pollution problems. ' Group III - Limits for the components are based on amounts in excess of which higher than normal operating cost would be encountered because of additional fuel requirements or refining times. (01-D-515-007-7) Table IV Determination of Oxidized Coal SCOPE: This test is used as a quick method to determine the amount of oxidized coal present in a sample. Oxidized coal is soluble in caustic, and results in a brown solution. This discoloration is proportional to the amount of oxidized coal. It should be noted that according to coal petrography, this method will detect oxidized coal that is greater than 2 to 5 percent, depending upon the coal type. It has been Clairton's experience that coals with less than 80 percent transmittance will present coal handling problems. REAGENTS AND MATERIALS: 1. 1N NaOH 2. 20 percent solution of Tergitol TMN in ethanol 3. 250 ml beakers 4. #40 and #42 Whatman filter paper 5. 60° glass funnels 6. Spectrophotometer (Bausch & Lomb Spectronic 20) (Fisher Scientific #7-143-1) 7. 4Hot plate 8. 3/4-inch test-tube-type cells 9. 100 ml graduate cylinder 10. Thermometer PROCEDURE: 1. Add 1 gram of coal sample that has been prepared to 100 percent minus 60 mesh to 100 ml of 1 normal NaOH. 2. Add 1 drop of Tergitol 3. Stir the coal and caustic; then heat on a hot plate at 85 + 2°C for l hour. Periodically stir the slurry as the solution is heating. 4. Filter the slurry through #40 and 589 filter papers. This double filtration is done in one operation with the #40 paper on top of the 589 paper 5. Bring the volume of the filtrate to 80 ml, using distilled water. 6. Measure the percent transmittance at 520 nm, using a blank of 1N NaOH that has had the same treatment as the samples to set 100 percent transmittance. 7. Report results in percent transmittance. (01-D-515-007-7) Property Calculated Stability Factor Volume Change (+ expansion, - contraction) at 52 1b/ft? (833 Kg/m3) dry oven bulk density Coking Pressure, psi @ 52 lb/ft? (833 Kg/m3) dry oven bulk density Table Vv Coking Characteristics of Different Ranks of Coal High Volat {le-A Coal Classification Medium Volatile Low Volatile Rank Rank Rank Low High Low High Low High <35 35 to 58 40 to 65 50 to 65 50 to 65 20 to 60 -5 w-30 -10 two-25 0 to -14 -5 to +10 -2 wtl10 +4 to +30 <1.0 0.5 to 2,2 1.0 w 5.0 2.0 to 10.0 5.0 to 15.0 10 to 30 (01-D-515-007-7) etd et dh wo 1200 . vee mane parte #) = Df 14.000 Catonfievatve parameter &/ 10 39,000 a . High-wilsble C brrumenous ae | Low-volstite rs Med-un-volattte High-wehie A ( | SubStuninous 8 Anieacite | Sermanthrecte nal ee 1 Coa! end subbtuminoys A 2 Peers trata tan wo eb. asim Tytem, they ore on rminerat-macter-tree bess Uy te woper terlt of caloric veiue ter chess 6 and high-wotelle A beuwinove cost. Figure 1. Comparison of Class Numbers and Boundar y Lines of International System with Group Names and Boundary Lines of ASTM System (01-D-515-007-7) 50 Europe — —-— Vitrinites ———=— Whole Coals United States 40 ——————=- Whole Coals — “~~ ‘“ 3 30 6 L e ~ ~~ z fe ~ -_ ~ 5 20 ° > 10 0 9 1.0 1.5 2.0 2.5 Reflectance, 2 (in oil) Figure 2. Reflectance-Volatile Matter Relations for European and United States Vitrinites or Whole Coals in the Bituminous — aeaeree Range (After McCartney). (01-D-515-007-7) seactives/Inerts 21.0 18.0 15.0 12.0 9.0 6.0 Figure 3. Reactives, vol Z Vitrinoid Types Optimum Ratio of Reactives to Inerts for Each Vitrinoid Type (01-D-515-007-7) Volatile Matter Gieseler Fluidity, ddpm 360 400 440 480 520 Temperature, °C Figure 4, Gieseler Plastometer Curves for Coals of Different Rank (Compatible coals for blending for coke making should have overlapping plastic ranges.) (01-D-515-007-7) 150 140 130 120 110 100 wo °o 80 70 Hardgrove Index 60 50 40 30 20 10 Fusinite | Semifusinite and 4 Pseudovitrinite —_— Vitrinite = Mineral Matter = + Micrinite and Exinite 0 10 20 30 40 . 50 Volatile Matter, wt % (dry, mineral-matter-free) Figure 5, Hypothetical Grindabilities of Different Macerals and Total Mineral’ Matter Calculated from Coal Sample Assumed to Contain 100 Percent of a Particular Maceral or of Mineral Matter (After Shaung-Shii Hsieh) (01-D-515-007-7) Rank Index Figure 6. 50 40 30 20 10 Inerts, vol Z Relationship of Rank Index, Vitrinoid Type, and Inerts of Coal (01-D-515-007-7) 13 12 ll 10 NMWEUO~ OO oo Vitrinoid Type 7.0 Stability + ¢. Factor 5.0 4.0 Strength Index (Rank) 3.0 Composition-Balance Index Figure 7, Correlation Graph for Predicting ASTM Tumbler Stability and Blending Potential From Basic Petrographic Data (01-D-515-007-7) ASTM Stability Factor 65 60 55 50 40 O-100% 36 1 V- 408 | A 408% | | l 208 ellie Coking Rate - 1.03 in./hr 7 fe 30 Bulk Density - 52-53 lb/ft D- 768 248 12 in. Pilot Oven @-100% — — 14 in. Pilot Oven 25. 40 50 60 70 80 Legend Alabama Medium Volatile Elkhorn No. 3 Seam Illinois No. 6 Seam Beckley Low Volatile Pittsburgh Seam Low Volatile Utah High Volatile Low & Medium Volatile Pittsburgh Seam 90 100 Pulverization Level, % minus 1/8 in. Figure 8 Effect of Coal Pulverization Level on the Strength of the Resultant Coke (01-D-515-007-7) Stability Factor Factor Hardness 66 ! | 62 ra ' _———- s __— Tt. 58 <= 3 —_——--- > a Mo Sire ae _-— o-4-7- 77 ~---=-6 54 : Blend: 49% Docena, 41% Concord 10% Hamilton 50 : 76 O- 58% - 1/8 in. ! O- soe - 1/8 in. &-1008= 1/8 in. 72 50 52 Bulk Density, lbs dry coal per ft? Figure $ Effect of Bulk Density on Stability and Hardness Factors of Resultant Cokes. (01-D-515-007-7) ' “ Oo C) 70 { + 8 | | | ° | 3 O a 6 \ | 3 7 | \ : | | | + 2 | | 4 s i w | | OS 4 = | | A | = 65 A 3 & | A Q | 2 ea | \ | 66 —s + 3 — ar HL a ¥ jo™ cae — » 9 oo! \ ~ } 1 5 r aS 4 a Legend ~~ \ \ rs O- 65% Elkhorn Seam oY a 8 35% Pocahoftas Low Yolatile N\A ® n 55 @ - 50% Illinois No. 5 Seam : i = 50% Medium Volatile N \ | | & Q- 75% Illinois No. 5 Seam | “ 25% Pocahontas Low Volatile \ \! A- 75% Eagle \ 25% Pocahontas Low Volatile 50 0.60 Figure 1g Effect of 0.80 / Coking Rate, in. per hour Coking Rate on the Strength of the Resultant Coke (01-D-515-007-7) 1.00 70 30 60 ~ Stabilit fF 20 § a came 3 50 =F é a a ~ - Ff 10 : g an 40 > = Volume Change $ 2 Lo 2 a [ 0 Ss 20 “ 30 & E 3 ~ < Pressure o fF 10 . = c 20 : . 3 ~ vy 20 = r 5 10 r & 30 0 Le 0.4 a6 0.8 LO 1.2 1.4 1.6 1.8 2.0 Reflectance of Vitrinoids, 2 Figure ll. Relation of Reflectance of Vitrinoids to Coke Stability, Volume Change (Expansion and Contraction), and Pressure for Individual Coals (01-D-515-007-7) 40.00 35.71 31.43 27.14 22.86 Organic Inerts, % 18.57 14,29 0.80 94 1.09 1.23 1.33 1.51 1.66 1.80 Reflectance, 2 Figure 37, Relation of Vitrinoid Reflectance and Coal Organic Inerts to Volume Change in a Sole-Heated Oven for Individual Coals (52 1b/ft?, 833 kg/m? bulk density) (01-D-515-007-7) Organic Inerts, 2% 40.00 35.71 31.43 27.14 22586 18,57 14,29 10.00 0.97 Volume Change, 2% 1.03 1.08 1.14 1.19 1.25 1.30 Reflectance, 2 Figure B Relation of Vitrinoid Reflectance and Coal Organic Inerts to Volume Change in a Sole-Heated Oven for a Coal Blend (52 1b/£t3, 833 kg/m bulk density) (01-D-515-007-7) 1.% 5.0 A = High-Volatile Coal B = Low-Volatile Coal 4.0 “A a a o 5 . 3.0 o ol a s 6 Sal oo 5 i 2.0 6 2 we 4 5 a 3 3 a 1.0 0 44 46 48 50 52 54 56 Dry Bulk Density, 1b/ft> Figure 14, Effect of Bulk Density on Coking Pressure for "AB" 2-Coal Blends (Coking Rate 1.0 inch, 25.4 mm per hour) (01-D-*15-007-7) 200.0 28.0 175.0 24.0 150.0 Bulk Dengity 7 20.0 55 lb/ft = 2 (881.0 kg/m? j| 125.0 8 ; 5 5 . ® 16.0 Bulk Density £ 50 1b/ft? , 100.0 a (800.9 kg/m”) M4 eo a ec x 12.0 | 75.0 oS Bulk Density x 45 1b/fe3 2 8.0 (720.8 kg/m3) 50.0 4.0 25.0 0 0 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 1.50 -60 1.70 1.80 1.90 Vitrinoid Reflectance (Ro), 2 Pigure 15, Relationship of Coking Pressure and Average Reflectance of Vitrinoids at Indicated Bulk Deasity (01-D-515-007-7) 700.0 Coking Pressure, Kn/m2 100 = 600.0 80 3 J 500.0 a . 3 s °° 400.0 be Be 6 e x — 300.0 8 40 1 200.0 20 “ 100.0 0 0 1.40 1.45 1.50 1.55 1.60 1.65 1.70 1.75 Vitrinoid Reflectance (Ro), % Figure 16 Relationship Between Coking Pressure and Petrographic Composition of Low-Volatile Coal (Bulk Density 51 1b/ft3, 817 kg/m3) (after Thompson et al) (01-D-515-007-7) PROCEDURES FOR CALCULATING COKE STRENGTH Work Unit I 1. Perform a maceral analysis. - Convert maceral count to percentage composition. . Determine ash and sulfur content of sample. - Calculate weight % mineral matter from ash and sulfur and convert to volume %. - Adjust percentage composition of sample (Item 2 above) to include both maceral and mineral material. oa fF WwW PP Work Unit II 1. Perform a reflectance analysis. 2. Convert raw values to: a. Mean maximum reflectance b. Percentages of each vitrinoid type. 3. Convert V-type percentages (total = 100) to actual Percentage composition in the sample (total - total vitrinoid percentage). Work Unit III 1. Transfer composition data to Coke Strength Master Work Sheet, apportioning 1/3 of the semifusinoids to the reactive category and 2/3 to the inert category, and totaling the reactive and inert maceral contents. 2. Assign all reactive maceral material to V-type class and total reactives by V-type. 3. Determine Strength Factors for each V-type from graph or tables. 4. Determine Optimum Inert Index and Balance Index in space provided on Master Work Sheet. 5. Determine Optimum Strength and Strength Index in space provided on Master Work Sheet. 6. Determine ASTM Stability Factor from Balance and Strength Indices, using either graph or tables. Coal Research Section The Pennsylvania State University Coke Strength Prediction Predict the ASTM Stability of Coke prepared from a coal for which the following petrographic data has been obtained: Vol. % mineral-free Total Vitrinoids 76.5 Total Exinoids 2.6 Total Fusinoids 11.1 Total Semifusinoids 2.8 Massive Micrinite 0.7 Granular Micrinite 6.3 100.0 V-Type Table V-Type Counts % of Total 10. - 4 2.0 11 134 67.0 12 62 31.0 The dry basis ash and sulfur contents are 5.10% and 0.82% respectively. MACERAL ANALYSES DATA SHEET | CoP lis SAMPLE IDENTIFICATION DATE. SOURCE OF SAMPLE OPERATOR PETROGRAPHIC COMPOSITION PELLET NO. VITRINITE PSEUDOVITRINITE SEMI FUSINITE FUSINITE MACRINITE MICRINITE SCLEROTINITE ALGINITE FLUORINITE EXUDATINITE BITUMINITE RESINITE OR PERCENTAGE CONVERSION TO PERCENTAGES EXCLUSIVE OF MINERAL MATTER CONVERSION TO PERCENTAGES INCLUSIVE OF MINERAL MATTER % TOTAL SULFUR 2.0 3.0 4.0 5.0 6.0. 7.0 8.0 % ASH (DRIED BASIS) MINERAL MATTER MODIFIED PARR’S FORMULA 9.0 10.0 11.0 12.0 13.0 XJQNI° \vUy 2 oO WwW =a > Er > INERTS, vol % Strength Index 5.0 4.0 3.0 {joj oj of it 10 08 0.6 Inert Rich <+>Inert Deficient Balance Index STABILITY FACTOR 65 REAv .‘VES/ INERTS 75.0 A REACTIVES vol % VITRINOID——> TYPES 3 ~<—REFLECTANCE, % 04 06 O8 10 #12 #14 #216 #18 20 22 COAL RESEARCH SECTION maa STRENGTH MASTER WORK SHEET THE PENNSYLVANIA STATE UNIVERSITY PSU SAMPLE NO. VITRINOID TYPES tn a foot a a TOTALS Sa] Tote! Reectivs mil o [ dad bas For Eoch Vitrinoid Type (sz) aa) ae) sx) (az) (sa) (ae) ae) aa) lar) bss )+( az) az) Cas) (as) (ia) * (iz) ‘haz ) ‘(aa) Ie led Total Inevts, Percent], Balonce 23.4 Optimum Inert Index ETE OE Hess WPsaeliTe ms FeROnID sud ean ST MTT EERIE CMTS aH IRICNPIMER SIT HALA ( PARTE pees ecceene Optimum Strength =| oom Strength Total Reoctives Index BALANCE INDEX STRENGTH INDEX MICROSTRENGTH STABILITY MEAN R